Riverside Audio to Text for Interviews

Riverside offers TXT for a reading copy and SRT for timed captions. For interviews, test whether the export can distinguish each speaker before uploading a full recording. Riverside documents single-speaker attribution in the standalone free uploader, so a multi-speaker interview needs a separate attribution check.

Worked example

Interview transcript review fixture

00:04  Host: Is the Cedar Lane pilot open to all ten teams?
00:09  Guest: Six teams are enrolled. Ten is the target.

Review: confirm Cedar Lane spelling, label both speakers, and keep the two numbers distinct.

Steps

  1. Take an approved sample with both speakers, one uncommon name and one number. Keep the recording's original timing for comparison.
  2. Export TXT for an article or transcript. If your deliverable includes captions, export SRT from the version of the edit you will publish.
  3. Check each change of speaker against the audio. Mark unclear words in the editorial draft and verify direct quotes by listening to the corresponding segment.

What to check

  • If the test transcript assigns both lines to one person, add speaker labels during review or choose an export workflow that can distinguish them.
  • After edits that change the running time, import the SRT into the final video and check a cue near the end.

See the Riverside listing

References